HEAL-T Code repository 


HEAL-T is a Matlab framework for the robust estimation of Heart-Rate (HR) from Blood-Volume-Pulse (BVP) signals and Accelerometer data in a noisy physical exercise environment. The framework is composed of ICA (channel decomposition/composition), moving average filters, and a dynamic variation of the bandwidth of a Blackman-Harris-Window (BHW) filter. The code was evaluated using the method described by the TROIKA framework, and with the TROIKA dataset posted in the ICASSP 2015 cup. This code allows the estimation of a robust and efficient HR regression using the re-calculation of a BHW filter bandwidth and a spectral peak searching and validation. For more details refer to the publication below.



Torres, J. M. M., Ghosh, A., Stepanov, E. A., and Riccardi, G. (2016). HEAL-T: An efficient PPG-based heart-rate and IBI estimation method during physical exercise. In Signal Processing Conference (EUSIPCO), 2016 24th European (pp. 1438-1442). IEEE


Compile the following form to download HEAL-T code.

Please provide your name and email address for your free download.